CRTITICAL_SERVICE_FAILED on startup disk check at 0%

Problem: Windows 10 needed to check my C (OS) drive for errors, so it scheduled a check and restarted. Upon boot, it went to the disk checking screen (showing the Windows logo and spinning dots), got to 0% and then immediately crashed with a CRITICAL_SERVICE_FAILED BSOD. This happened right after replacing my GPU (see backstory below) but that may be unrelated. How can I stop the disk check?

What I've tried: I've run chkdsk on that drive (and others) and found no errors. I've tried sfc /scannow but it doesn't work from the recovery command prompt. I've accessed the drive from another Windows install on a separate HDD and the files seem perfectly readable. I've tried safe mode - same result. I've tried replugging the SATA cable into a different port - no effect. I've tried "fsutil dirty query" and "chkntfs" and both show the disk as clean. I've also tried "chkntfs \x" to clear the dirty flag - no result.

Backstory: I recently had to replace my graphics card (Radeon HD 7800 replaced with GTX 950). Everything went fine and my computer restarted a few times. However, while diagnosing my graphics problems, I booted from my other drive (running Windows 7). Whenever this happens, Windows flags my drives as dirty and wants to check them. I'd been skipping the disk check at boot (at this point it have me 8 seconds to skip, rather than launching right in). But when everything was up and running, Windows still reported that it needed to check the disk, so I agreed and restarted. And then the above issue started. It's worth noting the I've had no disk problems except right before the disk check Steam did report trouble writing a downloading game to disk. It's also worth noting that while troubleshooting my graphics card my computer repeatedly crashed and had to be hard reset.

Goal: In the short term, I'd just like a way to disable the disk check, since the computer seems to be running ok and my only symptom of a faulty disk is dskchk failing... In the long term, obviously I'd like to fix the underlying issue and let windows check my disk.

I haven't run the dm_log_collector tool yet, since I can't boot to the affected OS, but I can do so on the other drive if that would be helpful. Alternately, if you can tell me what it collects, I can probably fetch it.

Thanks so much for any help!

Hi Ztruker, thanks for the quick response. As I said in the original post:

I've tried "fsutil dirty query" and "chkntfs" and both show the disk as clean. I've also tried "chkntfs \x" to clear the dirty flag - no result.

But I tried it again, this time from my Win7 command prompt (previously I had used the recovery console). Unfortunately, booting from Win7 is what convinced Win10 that the disk was dirty in the first place, so the "fsutil dirty query" did flag the drive as dirty this time. The ultimate result was that when booting to Win10 it tried a different kind of disk check ("Scanning disk for errors. This may take over an hour"), which actually completed successfully, and then rebooted to the same BSOD disk check at 0%.

Any other ideas? I think the disk check Win10 is trying, which was scheduled after a notification popped up in the action center asking me to do so, uses a different flagging system than the one fsutil and chkntfs interface with.

Well, after throwing the book at this problem, I finally tried switching the SATA cables for the SSD (failing) and the HDD (could boot), and now they both work. Hopefully it was just a loose cable or an incompatible port or some other hardware voodoo. Thanks for you help.